Socket Programming in C

Slides Adapted on

Jörn Altmann‘s Slides

CEN4500C 2

Questions that will be Addressed

What mechanisms are available for a programmer who writes network applications?How to write a network application that sends packets between hosts (client and server) across an IP network?Answer: socket API

Client ServerIP Network

CEN4500C 3

Socket ProgrammingTable of Contents1. Network Application Programming Interfa

ce: Sockets and Internet Sockets

2. Network Programming Tips3. Client-Server Architecture4. Example: Client Programming5. Example: Server Programming6. Network Programmer’s Mistakes

CEN4500C 4

Layers of the IP Protocol Suite

Link Layer

Transport Layer

Network Layer

Application Layer

Link Layer

Transport Layer

Network Layer

Application Layer

Ethernet

e.g. ftp

e.g. TCP, UDP

e.g. IP

CEN4500C 5

Protocol Suite Location Internet Protocol Layer

Link Layer

Transport Layer (TCP, UDP)

Network Layer (IP)

Application Layer

Network Card &Device Driver

(e.g. Ethernet card)

Operating System(e.g. Unix)

Applications(e.g. browser, game, ftp)

Application ProgrammingInterface (API)

(e.g. network API)

Interface to the Network Card

Location

CEN4500C 6

Network APIOperating system provides Application Programming Interface (API) for network applicationAPI is defined by a set of function types, data structures, and constantsDesirable characteristics of the network interface

Simple to use Flexible

independent from any application allows program to use all functionality of the network

Standardized allows programmer to learn once, write anywhere

Application Programming Interface for networks is called socket

CEN4500C 7

Sockets

Sockets provide mechanisms to communicate between computers across a network

There are different kind of sockets DARPA Internet addresses (Internet Sockets) Unix interprocess communication (Unix Sockets) CCITT X.25 addresses and many others

Berkeley sockets is the most popular Internet Socket runs on Linux, FreeBSD, OS X, Windows fed by the popularity of TCP/IP

CEN4500C 8

Internet Sockets

Support stream and datagram packets (e.g. TCP, UDP, IP)

Is Similar to UNIX file I/O API (provides a file descriptor)

Based on C, single thread model does not require multiple threads

CEN4500C 9

Types of Internet Sockets

Different types of sockets implement different communication types (stream vs. datagram)Type of socket: stream socket connection-oriented two way communication reliable (error free), in order delivery can use the Transmission Control Protocol (TCP) e.g. telnet, ssh, http

Type of socket: datagram socket connectionless, does not maintain an open

connection, each packet is independent can use the User Datagram Protocol (UDP) e.g. IP telephony

Other types exist: similar to the one above

CEN4500C 10

Network Programming Tips

Byte Ordering NamingAddressing

CEN4500C 11

Byte Ordering of Integers

Different CPU architectures have different byte ordering

D3

high-order byte low-order byte

memoryaddress A

memoryaddress A

+1

Stored at little-endian computer

Stored at big-endian computer

low-order byte high-order byte

F2Integer representation (2 byte)

CEN4500C 12

Message in Memory ofof big-endian Computer

Message is sent across Network 48 45 4C 4C 6F 01

00

Byte Ordering Problem

Question: What would happen if two computers with different integer byte ordering communicate?

Message is: [Hello,1]

Message is: [Hello,256]

Pro

cess

ing

48 45 4C 4C 6F 01 00

Message in Memory oflittle-endian Computer P

roce

ssin

g

Answer:Nothing if they do not exchange integers!But: If they exchange integers, they would get the wrong order of bytes, therefore, the wrong value!

Example:

CEN4500C 13

Byte Ordering SolutionThere are two solutions if computers with different byte ordering system want to communicate

They must know the kind of architecture of the sending computer(bad solution, it has not been implemented)

Introduction of a network byte order. The functions are:

uint16_t htons(uint16_t host16bitvalue)uint32_t htonl(uint32_t host32bitvalue)uint16_t ntohs(uint16_t net16bitvalue)uint32_t ntohs(uint32_t net32bitvalue)

Note: use for all integers (short and long), which are sent across the network

Including port numbers and IP addresses

CEN4500C 14

Network Programming Tips

Byte Ordering NamingAddressing

CEN4500C 15

Naming and Addressing

Host name identifies a single host (see Domain Name

System slides) variable length string (e.g. www.berkeley.edu) is mapped to one or more IP addresses

IP Address written as dotted octets (e.g. 10.0.0.1) 32 bits. Not a number! But often needs to be

converted to a 32-bit to use.

Port number identifies a process on a host 16 bit number

CEN4500C 16

Client-Server Architecture

Client requests service from serverServer responds with sending service or error message to client

Client Server

request

response

CEN4500C 17

Simple Client-Server Example

Client Serverrequest

response

socket()connect()send()

recv()close()

socket()bind()listen()accept()

recv()

send()

recv()close()

Connectionestablishment

Data response

Data request

End-of-file notification

CEN4500C 18

Example: Client Programming

Create stream socket (socket() )Connect to server (connect() )While still connected: send message to server (send() ) receive (recv() ) data from server and

process it

Close TCP connection and Socket (close())

CEN4500C 19

socket(): Initializing SocketGetting the file descriptor

int chat_sock;if ((chat_sock = socket(AF_INET, SOCK_STREAM, 0)) < 0) {

perror("socket"); printf("Failed to create socket\n"); abort ();

}

1.parameter specifies protocol/address family2.parameter specifies the socket type

Other possibilities: SOCK_DGRAM

3.parameter specifies the protocol. 0 means protocol is chosen by the OS.

CEN4500C 20

IP Address Data Structurestruct sockaddr_in {

short int sin_family; // Address family

unsigned short int sin_port; // Port number struct in_addr sin_addr; // Internet address

unsigned char sin_zero[8];

};

struct in_addr {

unsigned long s_addr; // 4 bytes

};

Padding of sin_zeros: struct sockaddr_in has same size as struct sockaddr

CEN4500C 21

connect(): Making TCP Connection to Serverstruct sockaddr_in sin;

struct hostent *host = gethostbyname (argv[1]);unsigned int server_address = *(unsigned long *) host->h_addr_list[0];unsigned short server_port = atoi (argv[2]);

memset (&sin, 0, sizeof (sin));sin.sin_family = AF_INET;sin.sin_addr.s_addr = server_address;sin.sin_port = htons (server_port);

if (connect(chat_sock, (struct sockaddr *) &sin, sizeof (sin)) < 0) { perror("connect"); printf("Cannot connect to server\n"); abort();}

CEN4500C 22

send(): Sending Packets

int send_packets(char *buffer, int buffer_len) { sent_bytes = send(chat_sock, buffer, buffer_len, 0); if (send_bytes < 0) { perror (“send"); } return 0;

}

Needs socket descriptor,Buffer containing the message, andLength of the messageCan also use write()

CEN4500C 23

Receiving Packets:Separating Data in a Stream

Use records (data structures) to partition the data stream

B

Fixed length record

Fixed length record

0 1 3

C

2 94 6 87

D

5

A

receive buffer

slide through

CEN4500C 24

Receiving Packetsint receive_packets(char *buffer, int buffer_len, int *bytes_read){

int left = buffer_len - *bytes_read; received = recv(chat_sock, buffer + *bytes_read, left, 0); if (received < 0) { perror (“recv"); } if (received <= 0) { return close_connection(); } *bytes_read += received; while (*bytes_read > RECORD_LEN) { process_packet(buffer, RECORD_LEN); *bytes_read -= RECORD_LEN; memmove(buffer, buffer + RECORD_LEN, *bytes_read); } return 0;}

Can also use read()

buffer

*bytes_read

buffer_len

CEN4500C 25

Server Programming: Simple

Create stream socket (socket() )Bind port to socket (bind() )Listen for new client (listen() )While

accept user connection and create a new socket (accept() )data arrives from client (recv() )data has to be send to client (send() )

CEN4500C 26

bind(): Assign IP and Portstruct sockaddr_in sin;

struct hostent *host = gethostbyname (argv[1]);unsigned int server_address = *(unsigned long *) host->h_addr_list[0];unsigned short server_port = atoi (argv[2]);

memset (&sin, 0, sizeof (sin));sin.sin_family = AF_INET;sin.sin_addr.s_addr = server_address;sin.sin_port = htons (server_port);

if (bind(chat_sock, (struct sockaddr *) &sin, sizeof (sin)) < 0) { perror("bind"); printf("Cannot bind server application to network\n"); abort();}

CEN4500C 27

bind():bind() tells the OS to assign a local IP address and local port number to the socket.Many applications let the OS choose an IP address.

Use wildcard INADDR_ANY as local address in this case.

At server, user process must call bind() to assign a portAt client, bind() is not required since OS may assign available port and IP address

The server will get the port number of the client through the UDP/TCP packet header

Note: Each application is represented by a server port number

CEN4500C 28

listen(): Wait for Connections

int listen(int sockfd, int backlog);

Puts socket in a listening state, willing to handle incoming TCP connection request.Backlog: number of TCP connections that can be queued at the socket.

CEN4500C 29

Server Example#define MYPORT 3490 // the port users will be connecting to #define BACKLOG 10 // how many pending connections queue will hold

int main(void) { int sockfd, new_fd; // listen on sockfd, new connection on new_fd struct sockaddr_in my_addr; // my address information struct sockaddr_in their_addr; // connector's address information int sin_size;

if ((sockfd = socket(AF_INET, SOCK_STREAM, 0)) == -1) { perror("socket");

exit(1); }

my_addr.sin_family = AF_INET; // host byte ordermy_addr.sin_port = htons(MYPORT); // short, network byte order my_addr.sin_addr.s_addr = INADDR_ANY; // auto. filled with local IP memset(&(my_addr.sin_zero), '\0', 8); // zero the rest of the struct

CEN4500C 30

if (bind(sockfd, (struct sockaddr *)&my_addr, sizeof(struct sockaddr)) == -1) { perror("bind"); exit(1);

}

if (listen(sockfd, BACKLOG) == -1) { perror("listen"); exit(1);

}

while(1) { // main accept() loop sin_size = sizeof(struct sockaddr_in); if ((new_fd = accept(sockfd, (struct sockaddr *)&their_addr, &sin_size)) == -1)

{ perror("accept"); continue;

}

printf("server: got connection from %s\n", inet_ntoa(their_addr.sin_addr));

if (send(new_fd, "Hello, world!\n", 14, 0) == -1) perror("send");

close(new_fd); }return 0;

}

CEN4500C 31

Client Example#include <netinet/in.h> #include <sys/socket.h>

#define PORT 3490 // the port client will be connecting to #define MAXDATASIZE 100 // max number of bytes we can get

// at once int main(int argc, char *argv[]) {

int sockfd, numbytes; char buf[MAXDATASIZE]; struct hostent *he; struct sockaddr_in their_addr; // server's address information

if (argc != 2) { fprintf(stderr,"usage: client hostname\n"); exit(1);

}

if ((he=gethostbyname(argv[1])) == NULL) { // get the host info perror("gethostbyname"); exit(1);

}

if ((sockfd = socket(AF_INET, SOCK_STREAM, 0)) == -1) { perror("socket"); exit(1);

}

CEN4500C 32

their_addr.sin_family = AF_INET; // host byte order their_addr.sin_port = htons(PORT); // short, network byte order their_addr.sin_addr = *((struct in_addr *)he->h_addr); // already network byte order memset(&(their_addr.sin_zero), '\0', 8); // zero the rest of the struct

if (connect(sockfd, (struct sockaddr *)&their_addr, sizeof(struct sockaddr)) == -1){ perror("connect"); exit(1);

}

if ((numbytes=recv(sockfd, buf, MAXDATASIZE-1, 0)) == -1) { perror("recv"); exit(1);

}

buf[numbytes] = '\0'; printf("Received: %s",buf); close(sockfd); return 0;

}

CEN4500C 33

I/O Blocking

socket();bind() ;listen();

whileaccept();recv() ;send() ;

Simple server has blocking problem Suppose 5 connections accepted. Suppose next accept() blocks. Other connections cannot send and receive. Cannot get keyboard input either.

CEN4500C 34

select() :I/O Multiplexingwaits on multiple file descriptors and timeoutreturns when any file descriptor is ready to be read or written or indicate an error, or timeout exceeded

advantages simple application does not consume CPU cycles while

waiting

disadvantages does not scale to large number of file

descriptors

CEN4500C 35

Example: Server Programming

create stream socket (socket() )Bind port to socket (bind() )Listen for new client (listen() )While Wait for (select() )

(depending on which file descriptors are ready)accept user connection and create a new socket (accept() )data arrives from client (recv() )data has to be send to client (send() )

CEN4500C 36

Server: Alternative Ways of Handling Many Clients

Forking a new process for each client: fork()But, creating new process is expensive.Multithreaded implementation: have one thread handling each client.Thread is like a process but light-weighted.

CEN4500C 37

Network Programmer’s Mistakes

byte orderingseparating records in streamsuse of select()misinterpreting the project specificationnot knowing all available system calls

CEN4500C 38

There are more System Calls

Depends on communication type Datagram sockets use recvfrom() and

sendto() for receiving and sending data

Closing connection: close(), shutdown()Convenient functions (on UNIX) inet_aton, inet_ntoa inet_pton, inet_ntop

CEN4500C 39

Literatureshort tutorial: Beej's Guide to Network Programming http://www.ecst.csuchico.edu/~beej/guide/net/

Unix Network Programming, volumes 1 and 2 by W. Richard Stevens. Published by Prentice Hall; ISBNs for volumes 1 and 2: 013490012X, 0130810819.Advanced Programming in the Unix Environment by W. Richard Stevens. Published by Addison Wesley. ISBN 0201563177. man pages on a Unix computer